Drekonja DM, Zarmbinski B, Johnson JR. Preoperative Urine Cultures at a Veterans Affairs Medical Center. JAMA Intern Med. 2013;173(1):71-72. doi:10.1001/2013.jamainternmed.834
Author Affiliations: Minneapolis Veterans Affairs Health Care System, Minneapolis, Minnesota (Drs Drekonja and Johnson), and Department of Medicine (Drs Drekonja and Johnson) and School of Medicine (Ms Zarmbinski), University of Minnesota, Minneapolis.
The value of preoperative urine screening is unproven, except before urologic procedures, in which detection and treatment of asymptomatic bacteriuria is beneficial.1 Despite this, authors of multiple small case series advocate for screening before nonurologic procedures.2- 5 However, patients with detected bacteriuria may undergo further testing6 and, if prescribed antimicrobial drugs, can develop diarrhea, allergic reactions, and Clostridium difficile infection (CDI).7 In addition, treatment of bacteriuria can delay procedures and extend hospitalization. Accordingly, we reviewed the medical records of patients who underwent cardiothoracic, orthopedic, and vascular procedures to document (1) the frequency of preoperative culture (UC) use, (2) the frequency of consequent antimicrobial therapy, and (3) any effect of preoperative urine screening, or consequent antimicrobial therapy, on postoperative complications.
We identified all cardiothoracic, orthopedic, and vascular surgical procedures performed during FY2010 at the Minneapolis VA Medical Center, and abstracted any UC order or result during the 7 days before each procedure. Bacteriuria was classified as high count (≥100 000 colony-forming units [CFU]/mL) or low count (10 000-90 000 CFU/mL). Cultures with fewer than 10 000 CFU/mL or no growth were considered negative for bacteriuria. Antimicrobial drugs for urinary tract infection (UTI) were recorded from 7 days before until 30 days after surgery, as were clinical manifestations or health care provider diagnoses of UTI. Assessed complications included allergy, diarrhea, CDI, and surgical site infection (SSI), as inferred from health care provider notes, polymerase chain reaction detection of CDI plus diarrhea, and application of Centers for Disease Control and Prevention SSI criteria.8
Overall, 1934 procedures were performed in 1688 patients. Orthopedic procedures predominated (1291 in 1115 patients), followed by cardiothoracic procedures (331 in 314 patients) and vascular procedures (312 in 259 patients). Patients were predominantly male (96%), with a mean age of 61.8 years. Overall, 13% of procedures were followed by complications, most commonly SSI (8%), followed by diarrhea (3%), UTI (2%), and CDI and drug allergy (<1%).
A UC was obtained before 25% of procedures, with significant variation by service (cardiothoracic, 85%; vascular, 48%; orthopedic, 4%; P < .001), but not by procedure type. No patient had UTI manifestations. Bacteriuria was detected by 11% of UCs (54 of 489). Whereas detection of high-count bacteriuria (5% overall) varied by service (cardiothoracic, 1%; vascular, 8%; orthopedic, 13%; P < .001), detection of low-count bacteriuria (6% overall) did not (data not shown). Antimicrobial drugs were dispensed for preoperative UTI in 16 patients, for a median of 4 days (range, 1-14 days). Interestingly, 8 treated patients (50%) had only low-count bacteriuria or a negative test result for UC.
To identify correlates of preoperative UC use, patients with (n = 489) and without (n = 1445) preoperative UCs were compared. Screened patients were older (66.9 vs 60.0 years; P < .001), more frequently male (98% vs 95%; P = .002), and more likely to develop SSI (17% vs 4%; P < .001), diarrhea (6% vs 2%; P < .001), and CDI (0.6% vs 0%; P = .02).
Among patients with a preoperative UC, patients with bacteriuria (n = 54) and those without (n = 435) were compared for postoperative complications, both overall and in relation to antimicrobial therapy. Surgical site infection was similarly frequent among patients with bacteriurua vs those without (20% vs 16%; P = .56). In contrast, postoperative UTI was more frequent among patients with bacteriurua vs those without (9% vs 2%; P = .01) (Table). Rates of other complications (diarrhea, CDI, allergy) did not differ by UC result or bacteriuria treatment (data not shown).
Among 54 patients with a positive screening UC, treated and untreated patients were compared with identify possible benefits or harms of such treatment (Table). Paradoxically, a greater proportion of treated patients developed a SSI (45% vs 14%; P = .03). This effect was greatest among patients with high-count bacteriuria, with SSI occurring in 4 of 8 (50%) if treated vs 1 of 15 (7%) if untreated (P = .03). Postoperative UTI also was numerically more frequent among treated patients than untreated patients (18% vs 7%). Other complications did not vary in frequency with treatment (data not shown).
We found that preoperative UCs were ordered inconsistently, that findings were rarely positive for bacteriuria, and that bacteriuria, when detected, usually was not treated. In addition, preoperative UCs were associated with higher rates of SSI, diarrhea, and CDI, whereas bacteriuria, although associated with health care provider–diagnosed postoperative UTI, was not associated with SSI. Because these associations are derived from small samples in an observational study, they should be interpreted cautiously, recognizing the potential for confounding. Similarly, the finding that treating bacteriuria was associated with SSI may be confounded by factors that contributed to the decision to administer antimicrobial drugs.
To our knowledge, this study provides the first systematic assessment of the frequency of preoperative UCs. Moreover, with nearly 2000 procedures, it is the largest study to assess outcomes associated with such testing. Our findings document that treatment of preoperative bacteriuria is associated with no benefit. These findings suggest that, outside the context of a randomized clinical trial, preoperative screening for and treatment of asymptomatic bacteriuria should be avoided in patients undergoing cardiovascular, orthopedic, or vascular surgery procedures.
Correspondence: Dr Drekonja, Medical Service, Infectious Diseases Section (111F), Minneapolis Veterans Affairs Medical Center, 1 Veterans Dr, Minneapolis, MN 55417 (firstname.lastname@example.org).
Published Online: December 3, 2012. doi:10.1001/2013.jamainternmed.834
Author Contributions: Dr Drekonja had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Drekonja and Johnson. Acquisition of data: Drekonja and Zarmbinski. Analysis and interpretation of data: Drekonja and Johnson. Drafting of the manuscript: Drekonja. Critical revision of the manuscript for important intellectual content: Drekonja, Zarmbinski, and Johnson. Statistical analysis: Drekonja. Obtained funding: Drekonja. Administrative, technical, and material support: Drekonja. Study supervision: Drekonja and Johnson. Final approval of manuscript: Drekonja, Zarmbinski, and Johnson.
Conflict of Interest Disclosures: Dr Johnson has research grants or contracts with Merck, Rochester Medical, and Syntiron.
Funding/Support: The study was supported by the resources of the Minneapolis Veterans Affairs Health Care System, including the Center for Epidemiological and Clinical Research and the Center for Chronic Disease Outcomes Research. Ms Zarmbinski has received scholarship support from the Infectious Diseases Society of America Medical Scholars Program.
Additional Contributions: Briana Ludtke, BA, and Kristina Poss provided assistance in conducting the study.